DecentMakeover comments

Results 43 comments of


                                            DecentMakeover

models

look for this `tf.app.flags.DEFINE_integer('batch_size_per_gpu', 8, '')` in multigpu_train.py and change accordingly

Unable to build docker for Cuda 8.0

were you able to run on cuda 8?

It is not changed by parameter in config.lua!

Hi Im trying to run the demo.py file, but im not able to find out where is the config.lua file? Am i missing something?

Error with running examples/text/demo.py

Hi Did you fix this , facing the same issue. Thanks

When do you release the code

Right here waiting...

CSV file used in training_script.py

@lunana123, @ifcoxs can you tell the difference between the label(y_train) and pai_id(PAI_ID). Thanks

Could you tell me your trained model is trained by which network structure?

@leacornelio Have you downloaded the model?

Could you tell me your trained model is trained by which network structure?

Btw, the software it is asking you to download is the baiduclient disk. Once you install it, you will be able to download the model. If you have succeeded in...

Few things to understand

@jaideep11061982 please close this issue if @MikkelAntonsen answer clears your doubts.

How to use only one gpu to train?

@Hung-Hsiang-617 I don't think any modifications are required to run this on a single GPU, I am running it on a single GPU and it works fine.